Methodological analysis demonstrates that sharp DIDISC and DISCID identify average treatment effects under continuous trends and no anticipation, clarifying causal estimation requirements.
This note revisits identification for difference-in-discontinuities (DIDISC) and discontinuity-indifferences (DISCID) when treatment assignment is sharp. I show that both estimands equal the average treatment effect at the cutoff (ATC) under two assumptions. The first assumption is the continuous trends assumption. The second assumption is no anticipation at the cutoff.
No takes yet. Share an insight, caveat, or question.
Geert Goeyvaerts (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: